Vaccine preventable invasive bacterial diseases in Italy: A comparison between the national surveillance system and recorded hospitalizations, 2007-2016.
Adolescent
Adult
Bacterial Infections
/ epidemiology
Child
Child, Preschool
Epidemiological Monitoring
Female
Haemophilus Infections
/ epidemiology
Hospital Records
/ statistics & numerical data
Hospitalization
/ statistics & numerical data
Humans
Incidence
Infant
Infant, Newborn
Italy
/ epidemiology
Male
Meningitis, Haemophilus
/ epidemiology
Meningitis, Meningococcal
/ epidemiology
Meningitis, Pneumococcal
/ epidemiology
Meningococcal Infections
/ epidemiology
Middle Aged
Patient Discharge
/ statistics & numerical data
Regression Analysis
Young Adult
Haemophilus influenzae
Hospital discharge records
Invasive bacterial disease
National surveillance system
Neisseria meningitidis
Streptococcus pneumoniae
Journal
Vaccine
ISSN: 1873-2518
Titre abrégé: Vaccine
Pays: Netherlands
ID NLM: 8406899
Informations de publication
Date de publication:
03 01 2019
03 01 2019
Historique:
received:
23
07
2018
revised:
24
10
2018
accepted:
15
11
2018
pubmed:
28
11
2018
medline:
7
6
2019
entrez:
28
11
2018
Statut:
ppublish
Résumé
Vaccine-preventable invasive bacterial diseases (IBDs) caused by Neisseria meningitidis (Nm), Streptococcus pneumoniae (Sp), and Haemophilus influenzae (Hi) have been notified in Italy since 2007 without assessing reporting completeness. Our study compared the number of cases of IBDs identified from the Italian Hospital Discharge Records (HDRs), using specific diagnostic ICD-9-CM codes, with those notified to the National Surveillance System (NSS) from 2007 to 2016. A multinomial logistic regression model was used to impute the aetiology of all discharges with a diagnosis of unspecified bacterial meningitis. Over a 10-year period, 14,243 hospital discharges with diagnosis of IBD were estimated in Italy (12,671 with specified aetiology and 1,572 with imputed aetiology). Among those, 2,513 (17.6%) were caused by Nm, 10,441 (73.3%) by Sp, and 1289 (9.1%) by Hi. Most invasive meningococcal diseases were coded as meningitis (72.3%), while Hi and Sp were more frequently coded as septicaemia (51.6% and 60.4%, respectively). The highest mean annual incidence rate was found for IBD caused by Sp (1.74 per 100,000), followed by Nm (0.42 per 100,000) and by Hi (0.21 per 100,000). Comparing NSS with HDR data, we found an initially high underreporting of all IBDs, and particularly for Hi. Data from the two systems overlapped in more recent years, due to an improved reporting completeness. The increasing IBD incidence observed in NSS data was not confirmed by HDR data trends, although with pathogen-related differences with Hi cases rising in both data sources, suggesting that is mainly due to an improved disease notification rather than to a true incidence increase. Comparing surveillance data with other data sources is useful to better interpret observed trends of notifiable diseases.
Sections du résumé
BACKGROUND
Vaccine-preventable invasive bacterial diseases (IBDs) caused by Neisseria meningitidis (Nm), Streptococcus pneumoniae (Sp), and Haemophilus influenzae (Hi) have been notified in Italy since 2007 without assessing reporting completeness.
METHODS
Our study compared the number of cases of IBDs identified from the Italian Hospital Discharge Records (HDRs), using specific diagnostic ICD-9-CM codes, with those notified to the National Surveillance System (NSS) from 2007 to 2016. A multinomial logistic regression model was used to impute the aetiology of all discharges with a diagnosis of unspecified bacterial meningitis.
RESULTS
Over a 10-year period, 14,243 hospital discharges with diagnosis of IBD were estimated in Italy (12,671 with specified aetiology and 1,572 with imputed aetiology). Among those, 2,513 (17.6%) were caused by Nm, 10,441 (73.3%) by Sp, and 1289 (9.1%) by Hi. Most invasive meningococcal diseases were coded as meningitis (72.3%), while Hi and Sp were more frequently coded as septicaemia (51.6% and 60.4%, respectively). The highest mean annual incidence rate was found for IBD caused by Sp (1.74 per 100,000), followed by Nm (0.42 per 100,000) and by Hi (0.21 per 100,000). Comparing NSS with HDR data, we found an initially high underreporting of all IBDs, and particularly for Hi. Data from the two systems overlapped in more recent years, due to an improved reporting completeness. The increasing IBD incidence observed in NSS data was not confirmed by HDR data trends, although with pathogen-related differences with Hi cases rising in both data sources, suggesting that is mainly due to an improved disease notification rather than to a true incidence increase.
CONCLUSIONS
Comparing surveillance data with other data sources is useful to better interpret observed trends of notifiable diseases.
Identifiants
pubmed: 30478004
pii: S0264-410X(18)31572-X
doi: 10.1016/j.vaccine.2018.11.047
pii:
doi:
Types de publication
Comparative Study
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
41-48Informations de copyright
Copyright © 2018 Elsevier Ltd. All rights reserved.